Parasitic Castration by the Digenian Trematode Allopodocotyle Sp. Alters Gene Expression in the Brain of the Host Mollusc Haliotis Asinina

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

FEBS Letters 580 (2006) 3769–3774

Parasitic castration by the digenian trematode Allopodocotyle sp. alters gene expression in the brain of the host mollusc Haliotis asinina

Tamika Ricea, Elizabeth McGrawa, Elizabeth K. O’Briena, Antonio Reverterb, Daniel J. Jacksona, Bernard M. Degnana,* a School of Integrative Biology, University of Queensland, Brisbane, Qld. 4072, Australia b CSIRO Livestock Industries, Brisbane, Qld. 4067, Australia

Received 22 March 2006; revised 16 May 2006; accepted 25 May 2006

Available online 9 June 2006

Edited by Takashi Gojobori

schistosome trematode, Trichobilharzia ocellata [7,10,11], cas- Abstract Infection of molluscs by digenean trematode parasites typically results in the repression of reproduction – the so-called tration appears to be caused by induced changes in gene parasitic castration. This is known to occur by altering the expression of the neuroendocrine system that controls growth expression of a range of host neuropeptide genes. Here we ana- and reproduction [2]. lyse the expression levels of 10 members of POU, Pax, Sox and The neurosecretory system of molluscs consists primarily of Hox transcription factor gene families, along with genes encod- individual or small clusters of neurons distributed throughout ing FMRFamide, prohormone convertase and b-tubulin, in the the cerebral, pleural, pedal and visceral ganglia [12]. The brain ganglia of actively reproducing (summer), non-reproducing antagonism between growth and reproduction appears to be (winter) and infected Haliotis asinina (a vetigastropod mollusc). controlled by neurohormones synthesised by cells in these gan- A number of the regulatory genes are differentially expressed in glia [13]. Physiological changes occurring in both normal and parasitised H. asinina, but in only a few cases do expression pat- infected L. stagnalis correlate with changes in expression of terns in infected animals match those occurring in animals where reproduction is normally repressed. genes encoding neuropeptides, such as: (i) schistosomin, which 2006 Published by Elsevier B.V. on behalf of the Federation of inhibits the action of gonadotropic hormones in reproductive European Biochemical Societies. organs; (ii) neuropeptide Y, which appears to regulate reproduction and growth; and (iii) caudodorsal cell hormone, which Keywords: Digenean; Hox; Pax; POU; Transcription factor; is known to regulate egg-laying and accompanying behaviours Sox; Haliotis [2]. The activities of specific cell clusters in the ganglia with known roles in reproduction and growth are also affected by parasitosis [7,14,15]. Regulatory transcription factors are likely to act as key controllers in the normal physiological transi- 1. Introduction tions operating in molluscan ganglia [16], and coordinate the expression of gene batteries that are required for the mollusc Digenean trematodes are platyhelminth parasites with com- to obtain a reproductive or growing physiological state. plex lifecycles, typically involving at least three different hosts Haliotis asinina (Vetigastropoda) has a distinct and predict- including a definitive, a first intermediate and a second inter- able reproductive season that lasts for about six months in the mediate host [1]. The first intermediate host is almost always summer [17]. During this season the ovaries are full of mature a mollusc, with a given digenean species capable of infecting oocytes, while during the non-reproductive winter period, the only one or two mollusc species [1]. These strict associations female gonad consists of small numbers of immature oocytes have allowed specialised interactions to evolve, including par- and granular debris [26]. Approximately 2% of wild population asite-induced alterations in host behaviour, defence function, of H. asinina of Heron Island Reef are infected with an opeco- nutrition, metabolism and reproduction [2–4]. One common elid digenean in the genus Allopodocotyle [8]. The condition of outcome of digenean infection in molluscs is castration [5–8], the gonads of these parasitised abalone imitates that of the which can result in the incomplete or total disruption of host non-gravid winter gonads, with no direct structural damage gamete production and presumably allow for the redeploy- observed. In infected abalone, the sporocyst and cercarial ment of metabolites to the parasite [9]. stages inhabit the haemocoel suggesting that parasitic castra- In most cases castration occurs indirectly, either through the tion is likely to be because of effects to the physiology of the deprivation of essential nutrients [6] or by the interruption and host and not direct consumption of the gonad by the trema- utilisation of the host’s natural endocrine system [7,10,11].In tode. Infections of other host molluscs largely appear to follow the interaction between the snail, Lymnaea stagnalis, and the similar patterns of castration [5–7]. In this study, we have quantitatively assessed the expression of 10 transcription factor genes, along with FMRFamide, prohormone convertase *Corresponding author. Fax: +61 7 3365 1655. 2 (PC2), and tubulin genes in the cerebral and pleuropedal E-mail address: [email protected] (B.M. Degnan). ganglia of H. asinina that are either: (i) actively reproducing; (ii) parasitized and castrated; and (iii) not reproducing. The Abbreviations: ANOVA, one-way analysis of variances; PC2, prohormone convertase 2; QPCR, quantitative real-time reverse transcription transcription factor families targeted for quantification – polymerase chain reaction Hox, Pax, POU and Sox – are known to play key roles in

0014-5793/$32.00 2006 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies. doi:10.1016/j.febslet.2006.05.068 3770 T. Rice et al. / FEBS Letters 580 (2006) 3769–3774 development of a wide range of animals and in some cases mer abalone), when not reproducing (winter abalone) or when physiological processes such as growth [18,19]. These genes infected by the digenean trematode (parasitised). The infected are expressed in tissue-restricted patterns in H. asinina and in animals were all obtained in the summer, when they normally brain ganglia [16,20–23]. would be reproducing.

3.1. Expression of reference gene HasSox-C 2. Materials and methods HasSox-Cwas chosen as the reference gene as it is widely expressed during development and across adult tissues [16,20,21]. 2.1. Collection of samples However, QPCR revealed that the HasSox-C expression levels H. asinina were collected on Heron Island Reef Australia. Parasi- in the cerebral ganglia were 1.7-fold different between winter tised abalone were identified by inspecting the gonad region for obvi- and summer samples (Fig. 1). Expression levels in the pleuro- ous infestation of sporocysts as per [8]. Total RNA was extracted from individual dissected cerebral and pleuropedal ganglia from reproduc- pedal ganglia were higher and more variable, with winter sam- ing (summer) abalone, non-reproductive (winter) abalone or parasi- ples having the highest level of expression and parasitised tised summer abalone as described in [16]. samples showing a significantly lower transcript levels than either of the uninfected treatments. HasSox-C was employed 2.2. Quantitative real-time PCR as a reference despite the uneven expression across seasons cDNA was synthesised using 0.2 lg of RNA, 10 lM random hexa- and between normal and parasitised animals. The effect of mer primers (Promega) and MMLV-reverse transcriptase (Promega) these differences in HasSox-C transcript abundance, however, according to the manufacturer’s specifications. Gradient polymerase chain reactions (PCRs) were performed on all primer sets (oligonucleo- was mitigated by defining a set of correction factors for each tide primer sequences available upon request) to ascertain a uniformly ganglia. Expression in each of the parasitised ganglia was set suitable annealing temperature, using 0.01 lg cDNA per 20 ll reac- to 1 and the fold difference between parasitised and the other tion. The cycling conditions for all gradient PCRs were 96 C for samples was calculated. The correction factors were as follows: 2 min, 35 cycles of 94 C for 30 s, gradient annealing from 53 to cerebral ganglia summer = 0.78 and winter = 1.34; and pleuro- 68 C for 1 min, 72 C for 1 min followed by a final extension of 72 C for 10 min and a cooling to 25 C for 10 min. pedal ganglia summer = 2.79 and winter = 4.25 (Fig. 1; Table Quantitative real-time reverse transcription polymerase chain reac- 1). These correction factors were applied to the normalised tion (QPCR) was conducted on a LightCycler (Roche) using Quanti- abundance values to produce the standardised abundance tech SYBR Green PCR Mix (Qiagen). The master mix was values. optimised to contain 0.75· SYBR Green Mix, 1 lM per primer, and 0.01 lg cDNA per 20 ll reaction. cDNA samples made from differing amounts of initial RNA (0.2 lg, 0.4 lg, 0.6 lg and 1 lg) were run as 3.2. Structural and neuropeptide gene expression profiles standards in each run. Standards were run in triplicate and each sam- Overall, HasTub1, HasPC2, and HasFMRFa transcripts ple in triplicate for the control gene and the gene of interest. All runs were more abundant in both ganglia compared to transcripts included a negative control. For all analyses the cycling conditions were 95 C for 15 min, 32 cycles of 94 C for 30 s, 56 C for 1 min, encoding transcription factors (Fig. 2). HasTub1 and HasPC2 72 C for 1 min with a read temperature of 80 C each cycle. Melting were expressed at higher levels in the ganglia of non-reproduc- curve analysis was used to establish that only one product was ampli- ing animals in the winter, with 2.4-fold higher expression of fied per reaction and that the product was the expected amplicon. HasTub1 in the cerebral ganglion and approximately 20-fold higher expression of HasPC2 in both ganglia (Table 1). Al- 2.3. Data analysis tered expression in parasitised abalone for both genes tended Analysis of fluorescence data was conducted on LightCycler Soft- to match the changes that existed between reproducing and ware version 3.3P9. Once the data had been collected from the Light- Cycler Microsoft Excel was used to normalise the target genes’ non-reproducing H. asinina. HasTub1 and HasPC2 transcript expression relative to HasSox-C expression. Normalisation was carried abundance increased in parasitised animals relative to summer out using the average HasSox-C expression per each specific run so as to account for variability between runs. All analyses were conducted within a ganglion type. Once normalised the values were then corrected for the uneven expression of the HasSox-C gene by multiplying by a correction value derived from the fold difference between the average HasSox-C values per treatment over all runs. SPSS Student Version 11.0 was then used to perform either one-way analysis of variances (ANOVA) or the non-parametric equivalent the Kruskal–Wallis test where Levene’s test showed variances to be unequal. Post hoc comparisons between treatments were carried out using either parametric or non-parametric tests (Tukey’s test or Nemenyi’s test, respectively) where applicable. For post hoc comparisons the critical rejection value was adjusted using Bonferoni’s method from 0.05 to 0.016. If no homogenous subsets were created at a = 0.016 despite a significant difference being found using ANOVA or Kruskal–Wallis post hoc comparisons were re-run at a = 0.05. See Reverter et al. [24,25] for a detailed rationale behind the normalisation procedure and statistical approach described here.

3. Results Fig. 1. Mean relative abundance for HasSoxC transcripts in H. asinina ganglia ± 1 S.E. of the mean. (A) cerebral ganglia; (B) pleuropedal Using QPCR, we determined the relative abundance of gene ganglia. Average abundances estimated from 10 QPCR analyses of the transcripts in cerebral and pleuopedal ganglia dissected from cerebral ganglia and 12 analyses of the pleuropedal ganglia. Homo- H. asinina either in the middle of their reproductive cycle (sum- geneous groupings are signiﬁcant at conﬁdence level of a = 0.016. T. Rice et al. / FEBS Letters 580 (2006) 3769–3774 3771

Table 1 HasPax-258 in the pleuropedal ganglion was similar to that Ratio of standardised abundance level between transcript level in observed in the cerebral ganglion, with higher transcript levels cerebral and pleuropedal ganglia (CG and PPG) from H. asinina in observed in non-reproducing and parasitised abalone. In both different seasons and normal vs. parasitised H. asinina genes, expression in the winter was significantly higher than Gene (tissue) Summer: Summer: Winter: summer, with parasitised animals not being significantly differ- winter parasitised parasitised ent from either class of normal animals. ** HasSox-C (CG) 0.58 0.78 1.34 In the Hox genes analysed the level of expression in the cere- HasSox-C (PPG) 0.66** 2.79** 4.25** HasTub1 (CG) 0.42** 0.52** 1.23 bral ganglion was too low to be detected at 32 cycles of ampli- HasTub1 (PPG) 0.95 1.23 1.29 fication. Three of the four Hox genes studied – HasHox-1,-2 HasPC2 (CG) 0.05** 0.10 2.11 and -3 – had similar patterns of expression in the pleuropedal HasPC2 (PPG) 0.07** 0.05** 0.69 ganglion with transcript abundance in non-reproducing win- HasFMRFa (CG) 1.34 3.54 2.64 ter animals higher than in summer and parasitised ani- HasFMRFa (PPG) 1.13 0.91 0.81 HasPOU-II (CG) N/R N/R N/R mals, sometimes significantly (Fig. 2; Table 1). Changes in HasPOU-II (PPG) 0.47 1.33 2.86** HasHox-5 expression were opposite to that observed in the HasPOU-III (CG) 1.00 33.65 33.65 other Hox genes with significantly lower levels detected in HasPOU-III (PPG) 0.17 1.09 6.53 the winter animals. HasPOU-IV (CG) 0.27 0.29 1.05 HasPOU-IV (PPG) 0.11 0.05* 0.43 HasPax-258 (CG) N/D N/D 0.50 HasPax-258 (PPG) 0.17* 0.80 4.77 4. Discussion HasPax-6 (CG) 0.44** 0.30** 0.67 * HasPax-6 (PPG) 0.31 0.58 1.89 All sexually mature Haliotis asinina (Vetigastropoda) repro- HasHox-1 (CG) N/D N/D N/D HasHox-1 (PPG) 0.30 3.52 11.77* duce in summer in a highly predictable manner [17], except the HasHox-2 (CG) N/D N/D N/D 2% of individuals that are infected with an Allopodoctyle HasHox-2 (PPG) 0.59 1.18 2.01 (Digenea: Opeceolidae) [8]. H. asinina heavily infected with HasHox-3 (CG) N/D N/D N/D opecoelid parasites lack mature gametes or developed gonads. * * HasHox-3 (PPG) 0.39 0.90 2.27 The parasites are not located within the gonad but in the sur- HasHox-5 (CG) N/R N/R N/R HasHox-5 (PPG) 1.90** 1.07 0.56** rounding haemocoel and therefore unlikely to be causing direct mechanical damage. This mode of parasitic castration Stars indicate significance of the ratio: *, significant at a = 0.05; and **, significant at a = 0.016. Significance based on the appropriate para- has been observed in many other gastropods [8]. From these metric (Tuckey’s) or non-parametric (Nemenyi’s) post hoc test. N/D, observations we sought to determine if parasitosis affects gene not detected (HasPax-258 only not detected in summer); N/R, not expression in the anterior brain ganglia – cerebral and pleuro- recorded. pedal ganglia – known to control the release of neurohormones that regulate growth and reproduction in other reproductive animals in a manner significantly similar to win- molluscs [12,13] and whether these changes mimic the normal ter non-reproductive animals. In the cerebral ganglion, Has- transition from reproductive to non-reproductive states in the FMRFa transcription appeared to be higher in the summer abalone host. Specifically, comparisons made between parasi- relative to parasitised and winter animals but none of the dif- tised and normal reproductive H. asinina reveal parasite-in- ferences were statistically significant. As in HasTub1, Has- duced changes in host gene expression. Changes that match FMRFa expression in the pleuropedal ganglion was relatively those occurring in non-reproductive (winter) abalone would constant. strengthen the supposition that parasitosis manipulates the normal physiological antagonism that exists between growth 3.3. Transcription factor gene expression profiles and reproduction in molluscs [2,11,14,15]. However differences The expression of POU genes in the cerebral ganglion (ex- in gene expression in the ganglia of reproductive and non- cept HasPOU-II, which was not determined in this ganglion) reproductive animals may also reflect other ganglionic func- did not vary significantly between seasons or between normal tions in addition to regulating reproduction and growth. and parasitised animals (Fig. 2; Table 1). In contrast, Has- Infection of H. asinina by the opecoelid digenean trematode POU-IV transcripts were about 4-fold more abundant in win- induces marked and sometimes statistically significant changes ter and parasitised animals relative to summer reproducing in expression levels of the genes surveyed in this study. Tran- animals. In the pleuropedal ganglion, HasPOU-II and Has- script levels in 10 of the 13 genes investigated change by 20% POU-III had similar expression profiles, with higher transcript or more in at least one ganglion of infected abalone compared levels in non-reproducing winter animals compared to summer to uninfected reproductive summer abalone. There is no trend and parasitised animals. In HasPOU-II, expression in the win- in the pattern of change, with transcript abundances increasing ter pleuropedal ganglion was significantly higher than in par- and decreasing to differing extents in parasitised animals. Also asitised animals. HasPOU-IV transcript prevalence in the genes are not consistently up or downregulated in both ganglia pleuropedal ganglion increased in winter and parasitised aba- (e.g. HasSox-C is upregulated in the parasitised cerebral gan- lone relative to reproducing animals. The 20-fold difference glion but downregulated the pleuropedal ganglion). The most in transcript levels between summer and parasitised H. asinina significant changes in gene expression in parasitised abalone was significant. compared to reproductively active abalone are the down-regu- Both Pax genes analysed showed a similar trends in the cere- lation of HasSox-C in the pleuropedal ganglion, up-regulation bral ganglion, with summer expression significantly lower (too of HasPC2 and HasPOU-IV in the pleuropedal ganglion low to be detected in HasPax-258) than in winter and parasi- and up-regulation HasTub1 and HasPax-6 in the cerebral tised animals (Fig. 2; Table 1). Expression of HasPax-6 and ganglion. 3772 T. Rice et al. / FEBS Letters 580 (2006) 3769–3774

Fig. 2. Normalised and standardized gene expression levels in cerebral and pleuropedal ganglia. Genes are listed to the left. White bars, ganglia derived from reproducing abalone in the summer; grey bars, ganglia derived from non-reproducing abalone in the winter; black bars; ganglia derived from abalone in the summer that are infected with the opecoelid digenean parasite. Normalised abundances are relative to HasSoxC levels (Fig. 1) and standardised abundances account for variation in HasSoxC between normal and infected animals. The correction factors were as follows: summer cerebral ganglia: parasited cerebral ganglia = 0.78; winter cerebral ganglia: parasitised cerebral ganglia = 1.34; summer pleuropedal ganglia: parsitised pleuropedal ganglia = 2.79; winter pleuropedal ganglia: parasitised pleuropedal ganglia = 4.25. Homogeneous groupings are signiﬁcant at conﬁdence level of a = 0.016 or (0.05, indicated with a*). #, cerebral sample not run, , cerebral sample too low to be detected at 32 cycles. T. Rice et al. / FEBS Letters 580 (2006) 3769–3774 3773

Comparison of transcript levels in brain ganglia of reproduc- References ing H. asinina in the summer with non-reproducing H. asinina in the winter indicates that the expression of all genes analysed [1] Jousson, O., Bartoli, P. and Pawlowski, J. (1999) Molecular have changed by 30% or more in at least one ganglion. All idenitification of developmental stages in Opecoelidae (Digenea). Int. J. Parasitol. 29, 1853–1858. genes are expressed at higher levels in the winter, except for [2] de Jong-Brink, M., Bergamin-Sassen, J.M. and Soto, M.S. (2001) HasFMRFa and HasHox-5, which are downregulated in the Multiple strategies of chistosomes to meet their requirements in winter in the cerebral and pleuropedal ganglia, respectively, the intermediate snail host. Parasitology 123, S129–S141. and HasTub1 and HasFMRFa in the pleuropedal ganglion [3] Adamo, S.A. (2002) Modulating the modulators: parasites, neuromodulators and host behavioural change. Brain Behav. and HasPOU-III in the cerebral ganglion, which are relatively Evol. 60, 370–377. unchanged. The cause of consistent upregulation of transcripts [4] El-Ansary, A. (2003) Biochemical and immunological adaptation in the brain of animals in the winter is unknown. While these in schistosome parasitism. Comp. Biochem. Physiol. 136, 227–243. parts of the mollusc central nervous system contribute to the [5] Pearson, E.J. and Cheng, T.C. (1985) Studies on parasitic regulation of growth and reproduction, it is likely the cerebral castration – occurrence of a gametogenesis-inhibiting factor in extract of Zoogonus lasius (Trematoda). J. Invertebr. Pathol. 46, and pleuropedal ganglia also integrate immediate and seasonal 239–246. sensory inputs. Without knowing the precise cells expressing [6] Crews, A.E. and Yoshino, T.P. (1989) Schistosoma mansoni – these genes, it cannot be determined if the differential expres- effect of infection on reproduction and gonadal growth in sion observed reflects a role in regulation of reproduction Biomphalaria glabrata. Exp. Parasitol. 68, 326–334. [7] Coustau, C., Robbins, I., Delay, B., Renaud, F. and Mathieu, M. and growth or is an outcome of a change in environmental in- (1993) The parasitic castration of the mussel Mytilus edulis by the puts. Nonetheless, the significant seasonal differences in the trematode parasite Prosorhynchus squamatus: specificity and expression of all the genes encoding transcription factors, the partial characterization of endogenous and parasite-induced prohormone convertase 2 gene and the b-tubulin gene indi- anti-mitotic activities. Comp. Biochem. Physiol. A 104A, 229– cates the regulatory and physiological states of the abalone 233. [8] Lucas, T., O’Brien, E.K., Cribb, T. and Degnan, B.M. (2005) brain are conspicuously different in the summer and winter. Digenean trematodes infecting the tropical abalone Haliotis While we observe large differences in transcript levels when asinina have species specific cercarial emergence patterns that comparing sexually reproductive H. asinina with either non- follow daily or semilunar spawning cycles. Mar. Biol. 148, 285– reproductive or parasitised abalone, in general these changes 292. [9] Valderrama, K., Oliva, M., Campos, B. and Brown, D. (2004) are not equivalent. Nevertheless, HasTub1, HasPC2, Has- Parasitic castration of Eurhomalea lenticularis (Bivalvia:Veneri- POU-IV, HasPax-258 and HasPax-6 are all upregulated in dae) by a digenetic trematode: quantitative histological analysis. winter non-reproductive and infected abalone compared to Dis. Aquat. Org. 59, 151–158. summer reproductive abalone. The extent of upregulation is [10] Joosse, J. and Vanelk, R. (1986) Trichobilharzia ocellata – statistically similar in winter and infected animals for Has- physiological characterization of giant growth, glycogen deple- tion, and absence of reproductive activity in the intermediate snail Tub1, HasPax-258 and HasPax-6 in the cerebral ganglion host, Lymnaea stagnalis. Exp. Parasitol. 62, 1–13. and HasPC2 and HasPOU-IV in the pleuropedal ganglion. [11] Hoek, R.M., VanKesteren, R.E., Smit, A.B., DeJongBrink, M. Since both infected and non-reproductive H. asinina have go- and Geraerts, W.P.M. (1997) Altered gene expression in the host nads that lack advanced gametes, the correlation of ganglionic brain caused by a trematode parasite: neuropeptide genes are preferentially affected during parasitosis. Proc. Natl. Acad. Sci. gene expression with gonad status suggests that there is a USA 94, 14072–14076. causal relationship. Given the well-documented role of neuro- [12] Dorsett, D.A. (1986) Brains to cells: the neuroanatomy of selected endocrine cells within ganglia in regulating growth, reproduc- gastropod species in: (Willows, A.O.D., Ed.), The Mollusca, Vol. tion and metabolism in a range of molluscs [12,13], it may be 9, pp. 101–187, Academic Press, Orlando, FL. that the upregulation of specific transcription factors activates [13] Geraerts, W.P.M., Smit, A.B., Li, K.-W., Hordijk, P.L. and Joosse, J. (1991) Molecular biology of hormones involved in the a cascade that results in changes in neuropeptide expression in regulation of reproduction and growth in molluscs. Bull. Inst. these cells, as been observed in both normal and infected mol- Zool. Acad. Sin. Monogr. 16, 387–440. luscs [2,11]. [14] Hordijk, P.L., Ebberink, R.H.M., Dejongbrink, M. and Joosse, J. In contrast, a number of gene expression levels in the pleu- (1991) Isolation of schistosomin, a neuropeptide which antago- nizes gonadotropic-hormones in a fresh-water snail. Eur. J. ropedal ganglion are more similar between reproductive sum- Biochem. 195, 131–136. mer and infected H. asinina compared to winter animals, [15] Schallig, H., Sassen, M.J.M. and Dejongbrink, M. (1992) In vitro with HasPOU-II, HasHox-1, HasHox-3 and HasHox-5 being release of the anti-gonadotropic hormone, schistosomin, from the significantly different. These may reflect seasonal differences central nervous system of Lymnaea stagnalis is induced with a in the regulation of the abalone’s physiology that are indepen- methanolic extract of cercariae of Trichobilharzia ocellata. Para- sitology 104, 309–314. dent of reproduction. [16] O’Brien, E.K. and Degnan, B.M. (2000) Expression of Pax258 in Pax-2 [27] and POU II [28] genes have been implicated in the gastropod statocyst: insights into the antiquity of metazoan immune responses in vertebrates. However, the abalone ortho- geosensory organs. Mar. Biotechnol. 2, 545–557. logues of these genes are not consistently up or downregulated [17] Counihan, R.T., McNamara, D.C., Souter, D.C., Jebreen, E.J., Preston, N.P., Johnson, C.R. and Degnan, B.M. (2001) Pattern, at this late stage of infection by the opoecoelid parasites. synchrony and predictability of spawning of the tropical abalone In summary, the overall pattern of gene expression in the Haliotis asinina from Heron Reef, Australia. Mar. Ecol. Prog. Ser. anterior ganglia of H. asinina heavily infected with the opeco- 213, 193–202. elid trematodes does not mimic that of a normally reproducing [18] Mangalam, H.J., Albert, V.R., Ingraham, H.A., Kapiloff, M., or non-reproductive animal suggesting that the parasite is Wilson, L., Nelson, C., Elsholtz, H. and Rosenfeld, M.G. (1989) Validation of alternative methods of data normalization in gene affecting more than just the control of reproduction. co-expression studies. Genes Dev. 3, 946–958. [19] Watkins-Chow, D.E. and Camper, S.A. (1999) How many Acknowledgement: This work was supported by an Australian homeobox genes does it take to make a pituitary gland? Trends Research Council grant to B.M.D. Genet. 14, 284–290. 3774 T. Rice et al. / FEBS Letters 580 (2006) 3769–3774

[20] O’Brien, E.K. and Degnan, B.M. (2002) Developmental expres- to account for heterogeneous variance across herds in carcass scan sion of a class IV POU gene in the gastropod Haliotisasinina traits from beef cattle. J. Anim. Sci. 75, 1477–1485. supports a conserved role in sensory cell development in bilate- [25] Reverter, A., Barris, W., McWilliam, S., Byrne, K.A., Wang, rians. Dev. Genes Evol. 212, 394–398. Y.H., Tan, S.H., Hudson, N. and Dalrymple, B.P. (2005) [21] O’Brien, E.K. and Degnan, B.M. (2002) Pleiotropic developmen- Validation of alternative methods of data normalization in gene tal expression of HasPOU-III, a class III POU gene, in the co-expression studies. Bioinformatics 21, 1112–1120. gastropod Haliotis asinina. Mech. Dev. 114, 129–132. [26] Jebreen, E.J., Counihan, R.T., Fielder, D.R. and Degnan, B.M. [22] Hinman, V.F., O’Brien, E.K., Richards, G.S. and Degnan, B.M. (2000) Synchronous oogenesis during the semilunar spawning (2003) Expression of anterior Hox genes during larval develop- cycle of the tropical abalone Haliotis asinina. J. Shellfish Res. 19, ment of the gastropod Haliotis asinina. Evol. Dev. 5, 508–521. 845–851. [23] O’Brien, E.K. and Degnan, B.M. (2003) Expression of Pax258 in [27] Chi, N. and Epstein, J.A. (2002) Getting your Pax straight: Pax the gastropod statocyst: insights into the antiquity of metazoan proteins in development and disease. Trends Genet. 18, 41–47. geosensory organs. Evol. Dev. 5, 572–578. [28] Ryan, A.K. and Rosenfeld, M.G. (1997) POU domain family [24] Reverter, A., Tier, B., Johnston, D.J. and Graser, H.U. (1997) values: flexibility, partnerships, and developmental codes. Genes Assessing the efficiency of multiplicative mixed model equations Dev. 11, 1207–1225.